An approach to adaptive steganography based on matrix embedding

Steganography, ("covered writing" in Greek) is the science of communicating in a hidden manner. The main requirement for a steganographic scheme is visual imperceptibility and statistical undetectability. Steganographic techniques which exhibit these two qualities have lesser embedding capacity. So as to achieve statistical undetectability with higher embedding efficiency, matrix embedding is preferred. But the research done so far suggests use of matrix embedding for improving embedding efficiency for a given approach, but this comes\´ at the cost of decrease in embedding capacity. In this paper so as to alleviate this problem we propose an approach based on matrix embedding. This approach gives higher embedding capacity while giving higher embedding efficiency. We make use of Weber-Fechner law to make the proposed method visually imperceptible. We think that this approach highlights the need for new methods, especially based on matrix embedding.